Applying artificial intelligence in data centres
Why is this gap important?
Demand for data centre services is expected to continue to grow strongly after 2020, and data centre energy use will continue to be largely determined by the pace of energy efficiency gains. While the continued shift to efficient cloud and hyperscale data centres will reduce the energy intensity of data centre services, applying artificial intelligence (AI) and machine learning to tap further efficiency gains may become increasingly important.
AI for data centre infrastructure management [TRL-8] can reduce the energy used by cooling infrastructure and power supply infrastructure.
AI for resource management [TRL-7] can improve servers workload optimisation, manage disk utilisation, and manage network congestion.
AI for data centre infrastructure management Readiness level:
AI for resource management Readiness level:
Colored bars represent the Technology Readiness Level (TRL) of each technology. Learn more about TRLs
What are the leading initiatives?
- Google has used DeepMind machine learning to their data centres to reduce energy use for cooling by up to 40%, and overall energy use of 15%. Building on this, Google is now applying AI to achieve a "tier-two automated control system" which adjusts cooling-plant settings automatically, in real-time. This could result in a further 30% decrease in energy use from cooling.
- Huawei released an AI-based "smart cooling" system in October 2018, which can reduce energy use by 8% and further optimise the energy consumption of data centres.
- Maya HTT has added machine learning capabilities in its data centre infrastructure management software to analyse servers and detect anomalies, and shift workloads to newer, more energy-efficient servers.